test: verify canonical styles in real playground flows

README.md

@@ -85,9 +85,10 @@ Each scenario creates a unique named daemon, disables native-comp JIT, loads
 the sibling repositories explicitly, creates one GUI frame, activates Emacs
 from the controlling shell, records through a PTY-backed macOS screen recorder,
 runs mount/resize/scroll/interaction checkpoints, writes screenshots and a
-manifest, then shuts down every process it created. Flex is always extracted
-from the committed `ebox-playground` HEAD, so a user's dirty fixture is never
-read, overwritten, staged, or restored.
+manifest, then shuts down every process it created. Flex and Grid are read from
+their current `ebox-playground` working files so the gate verifies the exact
+code a user is testing. The runner never writes, stages, restores, or otherwise
+mutates either fixture.
 
 A fresh capture intentionally reports `INCOMPLETE` until a human or agent has
 inspected `report.md`, `contact-sheet.png`, and its selected endpoint images.

README.zh-CN.md

@@ -81,8 +81,9 @@ scenario factory 的两个输入。增加新应用时不会复制第二套 daemo
 每个场景都会创建唯一命名 daemon，关闭 native-comp JIT，显式加载 sibling 仓库，
 创建单一 GUI frame，由外层 shell 激活 Emacs，通过保留 PTY 的 macOS recorder 录屏，
 依次执行 mount/resize/scroll/interaction checkpoint，生成 screenshot 与 manifest，最后
-清理自己创建的全部进程。Flex 始终从已提交的 `ebox-playground` HEAD 解包，绝不读取、
-覆盖、暂存或 restore 用户的脏 fixture。
+清理自己创建的全部进程。Flex 与 Grid 直接读取 `ebox-playground` 当前工作文件，确保
+门禁验证的就是用户正在测试的代码；runner 绝不写入、暂存、恢复或以其他方式修改
+这两个 fixture。
 
 新采集有意保持 `INCOMPLETE`，直到人或 agent 检查 `report.md`、
 `contact-sheet.png` 和报告选出的首尾图。完成时序审查后，对同一个 run directory

examples/research-shelf.el

@@ -699,7 +699,7 @@ slot has a reusable built-in fallback for direct composition."
   :view
   (column
          :class "research-shelf-shell" :width '(viewport)
-          :height 'auto :min-height '(viewport-height)
+          :height '(viewport-height)
           :color (etaf-theme-token :ink)
           :bgcolor (etaf-theme-token :paper)
    (slot :name 'header

scripts/run-gui-verification.sh

@@ -55,18 +55,8 @@ prepare_fixture() {
         research-shelf)
             SCENARIO_FIXTURE=""
             ;;
-        grid-reference)
-            SCENARIO_FIXTURE="$REPOS_DIR/ebox-playground/examples/grid-reference.ebox"
-            ;;
-        flex-reference)
-            SCENARIO_FIXTURE_ROOT="$SCENARIO_RUN_DIR/fixture"
-            mkdir -p "$SCENARIO_FIXTURE_ROOT"
-            git -C "$REPOS_DIR/ebox-playground" archive \
-                --format=tar --output="$SCENARIO_RUN_DIR/flex-fixture.tar" \
-                HEAD examples/flex-reference.ebox
-            tar -xf "$SCENARIO_RUN_DIR/flex-fixture.tar" \
-                -C "$SCENARIO_FIXTURE_ROOT"
-            SCENARIO_FIXTURE="$SCENARIO_FIXTURE_ROOT/examples/flex-reference.ebox"
+        flex-reference|grid-reference)
+            SCENARIO_FIXTURE="$REPOS_DIR/ebox-playground/examples/$SCENARIO_NAME.ebox"
             ;;
         *) usage ;;
     esac

tests/etaf-playground-tests.el

@@ -464,8 +464,8 @@ database and mounts a test buffer before running BODY."
      '(:viewport-width 900 :viewport-height 40))
     (let* ((state (ebox--buffer-render-state (get-buffer buffer)))
            (root (plist-get state :root-node)))
-      (should (eq (plist-get root :height) 'auto))
-      (should (equal (plist-get root :min-height) '(viewport-height)))
+      (should (equal (plist-get root :height) '(viewport-height)))
+      (should (= (plist-get root :min-height) 0))
       (should-not (plist-get state :scroll-region-ids)))))
 
 (ert-deftest etaf-playground-fixture-supports-realistic-page-counts ()